Carbon stocks and fluxes in Asia-Pacific mangroves: current knowledge and gaps

Despite more regional estimates for the offshore export of mangrove-derived carbon now becoming available, particularly in the past five or six years (80% of the global reports have materialized since 1995), a wide-scale dataset for the AP region has not been produced. Among the 25 AP countries considered, just 9 reported lateral carbon flux estimates; 45% of these came from Southeast Asian countries like Palau, Vietnam, Philippines, and Papua New Guinea (table 2). Because of insufficient quantitative estimates for mangrove-derived DOC, POC and DIC, it is impossible to generalize the role of AP mangroves as a carbon source or sink. This study confirms global estimates which found it common for AP mangroves to export DOC to the open sea. Globally, AP mangroves export the most amount of DIC, whereas exported POC was reported to be the greatest in Amazonian mangroves (mean 876.5 g C m⁻² yr⁻¹) (Dittmar et al 2006). Area-normalized yearly fluxes of DOC, POC and DIC differ considerably, depending on tidal settings and calculation method. Maximum DOC, POC and DIC exports were seen in the Indian Sundarbans (705 g C m⁻² yr⁻¹), Vietnam (2152 g C m⁻² yr⁻¹) and Papua New Guinea (285 g C m⁻² yr⁻¹) respectively, while minimum exports were reported in western Taiwan (for all forms ∼1 g C m⁻² yr⁻¹). High discharge from the river Ganges and the macrotidal areas of the Bay of Bengal induces a significant flushing of DOC and sediment-eroded POC (136 g C m⁻² yr⁻¹) away to the Bay of Bengal (Ray and Weigt 2018). With the exception of one study in Evan Head, Australia, where negative POC flux was reported (meaning import), all estimates revealed an export of DOC, DIC and POC, suggesting that mangroves play a role in 'carbon outwelling' (Santos et al 2021).

The level of uncertainty in carbon flux estimates in estuaries depends on the method used, with some methods resulting in very large uncertainties. Six different methods were employed for flux estimation across the 13 export flux measurements seen. This lack of methodological uniformity contrasts with more standardized methods used to calculate other flux components, like carbon burial (which uses a ²¹⁰Pb dating method), soil emission (chamber method) or water-air exchange (bulk formula); this could explain why export or outwelling flux estimations are very rare for mangrove ecosystems. This study highlights the need to establish a standard method to estimate flux globally, which would contribute to a more precise estimate of AP mangroves' blue carbon budget. Regardless of methods used, if we include non-AP mangroves and draw a global mean figure for the export/import of DOC, POC and DIC, final estimates differ significantly to those of (Alongi 2020); with this study giving 1156 g C m⁻² yr⁻¹, and the Alongi study estimating 2197 g C m⁻² yr⁻¹ (Alongi 2020); this is because the later review showed only positive export flux values and some findings were not up-to-date. Our estimates further reveal that AP exports of DOC, DIC and POC were 395 ± 20 (figure 2(a)), 102 ± 49 (figure 2(b)) and 659 ± 133 (figure 2(c)) g C m⁻² yr⁻¹, respectively.

Zoom In Zoom Out Reset image size

Despite being extensively explored for blue carbon research and having the maximum global coverage, Indonesian mangroves suffer from a complete lack of data on carbon export fluxes to date. This is particularly striking, as some of the highest values of SCSs, carbon accumulation and burial rates, and soil emissions are reported in Indonesia; but carbon loss via export, and carbon gain via import, have never been quantified. As such, the fate of carbon remains unknown, and a concrete blue carbon budget for Indonesian mangroves cannot be established. This is particularly true for East Kalimantan, where—except for carbon stock, burial and soil emission fluxes—there is no information on mangrove-derived DOC, POC and DIC exchange fluxes, contributing to the challenge of establishing a comprehensive budget. This lack of quantitative estimates for carbon outwelling should encourage researchers to initiate flux measurement in Indonesia.

The major limitations of present flux estimates are manifold, including the influence held by wide-ranging hydrological ecological processes. Recent literature has discussed such limitations, highlighting for example the presence of green carbon in the blue carbon pool, delivered from upstream and mixed with estuarine water (Ray et al 2021b). Stable isotopes alone are not helpful in separating blue carbon from green carbon sources, unless other methods—like the eDNA method Ortega et al (2020)—are developed and applied. This remains a knowledge gap in blue carbon studies and warrants more research using e-DNA in future.

Although mangroves occupy just 0.5% of global land area, they sequester around 25% of the total carbon accumulated by vegetated habitats like saltmarshes, seagrasses and tropical peat (Duarte et al 2005). We registered CAR estimates from 14 countries in the AP region; overall, more than the carbon export inventory (table 2). This is likely because a standard method exists for such measurements, based on a combination of sedimentary carbon and soil accumulation rates, estimated from radioisotopes (Sanders et al 2010). Most results were retrieved from Southeast Asian countries (18), followed by PO (9), East Asian (4) and South Asian (1) countries. Indonesia ranked highest in quantifying CAR from dated cores collected from its various mangrove locations. CARs rarely exceeded 200 g C m⁻² yr⁻¹ but this differed at the regional and country scale. Some of the highest CAR were reported for Indonesia (>500 g C m⁻² yr⁻¹) (Kusumaningtyas et al 2019), while Thailand, Philippines, Vietnam, Malaysia and Cambodia reported ∼200 g C m⁻² yr⁻¹. Overall ranges of CAR in EA, SEA and the PO subregions were 108–444 (206 ± 159), 50–1722 (321 ± 386) and 26–450 (176 ± 125) g C m⁻² yr⁻¹, respectively. There was no significant difference (F_{28, 30} = 0.730; P = 0.491) between CAR from EA, SEA and PO regions. In SA, CAR was reported only from the Indian Sundarbans (60 ± 17 g C m⁻² yr⁻¹). Overall, mean CAR for AP mangroves was estimated at 210 ± 153 g C m⁻² yr⁻¹ (excluding one extreme value that resulted in uncertainty >100%) (figure 2(d)); which is not much different from non-AP regions (233 ± 177 g C m⁻² yr⁻¹), but slightly higher than the global estimate of 162 g C m⁻² yr⁻¹ (Alongi 2020).

Although no significant difference in CAR was observed (based on a one-way ANOVA comparison, p > 0.05) across the diverse mangrove types (intact, restored and degraded), there was a clear trend in that interior intact mangroves had higher CAR than fringed and degraded mangroves, for example Bintuni Bay (Murdiyarso et al 2021) and northern New Zealand (Pérez et al 2017). The reduction of hydrological flushing in interior mangroves supports the accumulation of organic carbon-rich debris on the sediment surface (Krauss et al 2014). System-specific variabilities in allochthonous or autochthonous inputs (e.g. sedimentation and/or decomposition rates) may also cause such differences in CAR across the reported mangroves.

Besides radiometric dating, chrono-sequential observation—or space-for-time-substitution—offers an indirect approach or type of 'natural experiment'(Azman et al 2023), as applied to relatively younger sites in the Philippines (MacKenzie et al 2021). This kind of CAR estimations by radio tracers are based on the assumption that sediment and organic carbon accumulation occur steadily during the period of accumulation. The 'indirect way' (chrono-sequencing) cannot be applied to a steady-state system (i.e. matured or climax forest); only to relatively younger sites. As such, these two approaches are mutually exclusive and useful only for specific objectives.

Mangrove soil CO₂ release results were available for ten AP countries, with most available for SEA (table 2). South China and Hong Kong, and the Philippines generated the largest mean figures for CO₂ emissions (∼4110 g C m⁻² yr⁻¹), while the smallest mean emissions were recorded from North Sulawesi, Indonesia and Lothian, Sundarbans (∼240 g C m⁻² yr⁻¹). Overall mean CO₂ fluxes for EA, SEA, SA and PO were 2191 ± 1917, 1217 ± 1012, 931 ± 836, 1110 ± 1052 g C m⁻² yr⁻¹, respectively (figure 2(e)). There was no significant difference (F_{26, 29} = 1.08; P = 0.375) between soil CO₂ fluxes in EA, SEA, SA and PO. The most up-to-date average CO₂ flux for AP mangroves overall is 1350 g C m⁻² yr⁻¹, which is two times higher than the global mean of 613 g C m⁻² yr⁻¹ (Alongi 2020), suggesting that comparatively AP mangroves are a significant potential source of CO₂. Large uncertainties around results are linked to spatial heterogeneity across the study locations, as well as differences in the techniques applied for greenhouse gas measurement. An example of this system heterogeneity is the Indian Sundarbans, where two islands show significantly different values (Henry: 263; Lothian: 2117 g C m⁻² yr⁻¹). Most measurements of CO₂ flux at the sediment-air interface have been made using a custom-built system comprising of a chamber, either light or dark, enclosing a small area of the soil surface, preferably avoiding the numerous biogenic structures like pneumatophores (Kristensen et al 2008, Troxler et al 2015). Differences in mean CO₂ flux between light and dark chamber measurements are insignificant (one-way ANOVA, p> 0.05) which might be due to environmental factors as described by others (Coelho et al 2009, Leopold et al 2013), e.g. desiccation of surface sediment due to sun exposed daytime condition and increased evapotranspiration may have short term effects on the photosynthetic activity of microphytobenthos. However, removal of algal mat from the chamber reveals significant differences when compared to intact sediment (p< 0.005), suggesting that photosynthetic organisms play a vital role in benthic CO₂ uptake (Leopold et al 2013) Considering at edaphic factors, such as soil temperature, moisture and redox potential, these are known for governing the spatio-temporal variability of CO₂ fluxes (Chen et al 2012, Leopold et al 2013). The limited data that is available for planted and restored mangroves shows higher CO₂ fluxes (1837 g C m⁻² yr⁻¹) than in intact mangroves (1137 g C m⁻² yr⁻¹), but region-specific results—such as from Sulawesi, Indonesia—show practically no difference between mangrove types (Cameron et al 2019). Some of the highest rates of CO₂ emissions come from cleared or degraded mangroves (for example in Honda Bay, Philippines and northern New Zealand); this infers that simply preventing deforestation can serve as an excellent way to preserve threatened carbon stocks (Lovelock et al 2011).

In comparison to soil respiration, pelagic respiration studies are conducted at a limited scale in AP mangroves. Data on water-air CO₂ fluxes was obtained from surveys in six AP countries. We found the magnitude of air-water CO₂ fluxes in these countries was quite similar to that of soil CO₂ fluxes. Reports from SEA were limited to Indonesia and Vietnam (43–1486 g C m⁻² yr⁻¹), whereas Japan (58–1314 g C m⁻² yr⁻¹) and India (18–1570 g C m⁻² yr⁻¹) were the only countries to represent East and SA, respectively. In the PO, this data came from Australia (548–3928 g C m⁻² yr⁻¹) and Papua New Guinea (803 g C m⁻² yr⁻¹). Overall mean CO₂ flux from AP mangrove waters was 1007 ± 870 g C m⁻² yr⁻¹ (figure 2(f)). A bulk formula method is applied in most studies on AP mangroves; unlike in the Everglades, where a recently-developed dual tracer of SF₆ and ³He has been routinely applied over the last decade (Ho et al 2016). Our survey suggests that mangrove waters on the peripheries of Australian and Indian Sundarbans landmasses have received the most substantial attention with respect to quantifying air-water CO₂ fluxes. The lowest CO₂ flux—reported in the Sundarbans estuaries—is ascribed to greater dilution of organic carbon-poor seawater, with enhanced phytoplankton uptake resulting in lower water pCO₂ (Biswas et al 2004, Akhand et al 2021). In contrast, greater CO₂ flux from creek and river waters in other locations suggests higher input of land-derived organic carbon, leading to pCO₂ enrichment in water (Call et al 2015). Therefore, based on the type of estuary, i.e. semi-enclosed creeks or perennially-open bays (e.g. Bay of Bengal) and other factors (e.g. gas transfer velocity) fluxes might vary to a great extent.

The gross primary productivity (GPP) is an important component to understand the mangrove blue carbon cycle. In our analyses we did not include this component due to several reasons; (1) the scarcity of data in AP region during our study period from 2011 to 2020, (2) use of different methodologies to estimate the GPP parameters such as remote sensing (Kanniah et al 2021), tree diameter increment (Kamruzzaman et al 2019) and litterfall production (Sharma et al 2012), photosynthesis rate (Wongpattanakul et al 2015), and leaf area index based estimation of productivity (Clough et al 2000). Root productivity is a major concern to understand the mangrove belowground productivity. Long-term research is needed to monitor GPP under ongoing climate change (changing pattern in temperature and rainfall, and sea level rise) to understand how mangrove will act as a source or sink. Further research is needed to understand spatial and temporal GPP and net ecosystem productivity (NEP) across AP region to refine mangrove blue C cycle.

Mangrove stem mediated flux was not included in the current analysis. This is a new area of research field and currently most of the researchers are interested in it. Especially, it is still not clear how the stems from different species play roles in the carbon budget of mangrove ecosystems. Plant mediated CO₂ emission was done in the Florida coastal Everglades, USA (Troxler et al 2015), therefore similar can be done in AP region. According to available evidence tree-stem methane (CH₄) emissions may be an important and unaccounted component of carbon budgets (Jeffery et al 2019). Therefore, we did not include mangrove stem mediated carbon flux in our analysis's due scarcity of data in this context. Further, studies are needed to incorporate tree-stem mediated carbon budget into mangrove blue carbon cycle to understand the source or sink pattern in AP region.